ABSTRACT
The impacts of present and potential future climate change will be one of the most important scientic and societal challenges in the 21st century. Given observed changes in temperature, sea ice, and sea level, improving our understanding of the climate system is an international priority. is system is characterized by complex phenomena that are imperfectly observed and even more imperfectly simulated. But with an ever-growing supply of climate data from satellites and environmental
4.5 Climate Data Analysis: Problems and Approaches 91 4.5.1 Abrupt Changes 91 4.5.2 Climate Networks 94 4.5.3 Predictive Modeling: Mean Processes and Extremes 96
4.6 Seasonal Climate Forecasting 97 4.6.1 What Is the Basis for Seasonal Forecasting? 97 4.6.2 Data Challenges 99 4.6.3 Identifying Predictable Quantities 99 4.6.4 Making the Best Use of GCM Data 100
4.7 Climate Extremes, Uncertainty, and Impacts 101 4.7.1 e Climate Change Challenge 101 4.7.2 e Science of Climate Extremes 101 4.7.3 e Science of Climate Impacts 102
4.8 Reconstructing Past Climate 103 4.8.1 e Global Temperature Reconstruction Problem 104 4.8.2 Pseudoproxy Experiments 107 4.8.3 Climate Reconstructions and the Future 108
4.9 Applications to Problems in Polar Regions 110 4.10 Toward a Climate Informatics Toolbox 112 4.11 Data Challenges and Opportunities in Climate Informatics 114
4.11.1 Issues with Cross-Class Comparisons 114 4.11.2 Climate System Complexity 116 4.11.3 Challenge: Cloud-Computing-Based Reproducible
Climate Data Analysis 116 4.11.3.1 Data Scale 117 4.11.3.2 Reproducibility and Provenance Graphs 117
4.12 Conclusion 118 Acknowledgments 118 References 119
sensors, the magnitude of data and climate model output is beginning to overwhelm the relatively simple tools currently used to analyze them. A computational approach will therefore be indispensable for these analysis challenges. is chapter introduces the edgling research discipline climate informatics: collaborations between climate scientists and machine learning researchers in order to bridge this gap between data and understanding. We hope that the study of climate informatics will accelerate discovery in answering pressing questions in climate science.
